644871 - Combine MarkAndSweep/MarkAndSweepCompartment

Status: RESOLVED FIXED

(Core :: JavaScript Engine, defect)

Description

[Gregor Wagner [:gwagner]]

These functions are mostly redundant and it happened to me more than once that I only changed one function...

Comment 1

[Gregor Wagner [:gwagner]]

Attachment: patch

Comment 2

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

Great! Let's get this in soon.

One thing I noticed: when moving shapes to the GC heap, I forgot to take out the TIMESTAMP code for shape sweeping. It doesn't really time anything anymore. This might be a good opportunity to take it out.

Comment 3

[Gregor Wagner [:gwagner]]

(In reply to comment #2)
> Great! Let's get this in soon.
> 
> One thing I noticed: when moving shapes to the GC heap, I forgot to take out
> the TIMESTAMP code for shape sweeping. It doesn't really time anything anymore.
> This might be a good opportunity to take it out.

We could still time the shape finalization if we move it out of finalizeObjectArenaLists.


Brendan, is there some magic behind DumpLoopStats and DumpScopeDepthMeter? I can't find an implementation for them. Is it safe to remove?

Comment 4

[Gregor Wagner [:gwagner]]

Attachment: patch

Comment 5

[Gregor Wagner [:gwagner]]

Attachment: patch

I restored the shape-finalization timing again. 
Otherwise there shouldn't be any change to the previous code.

Comment 6

[Gregor Wagner [:gwagner]]

Attachment: patch

Now for real...

Comment 7

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

Comment on attachment 521882 [details] [diff] [review]
patch

Could you also take out the sentence about how "We finalize iterators before other objects..."? As far as I can tell, that's just not true any more.

Comment 8

[Gregor Wagner [:gwagner]]

http://hg.mozilla.org/tracemonkey/rev/5c7a2ced20f8

Comment 9

[Chris Leary [:cdleary] (not checking bugmail)]

http://hg.mozilla.org/mozilla-central/rev/5c7a2ced20f8